SOC 8311 Basic Social Statistics

1
Evolution of the Strategic Alliance Network in
the Global Information SectorDavid Knoke Xi
Zhu University of Minnesota SIENA
Workshop Groningen University January 8-11, 2007
Thanks to the Ford Foundation, Digital Media
Forum, and University of Minnesota for funding
and to Anne Genereux, Song Yang, and Francisco J.
Granadosfor research assistance.
2
Corporate Social Capital
Social Capital Resources accruing to an ego
actor through direct and indirect relations with
its alters that facilitate egos attainment of
its expressive or instrumental goals.
Diverse conceptualizations of an actors social
capital

• inheres in the structure of relations between
  persons and among persons (Coleman 1990302)
• at once the resources contacts hold and the
  structure of contacts in the network (Burt
  199212)
• resources embedded in a social structure which
  are accessed and/or mobilized in purposive
  action (Lin 200112)

Corporate Social Capital (CSC) Social relations
embedded in work-related organizational roles
(e.g., workers, teams, executives, owners), not
in their personal networks.
Corporate social capital, then, refers to The
set of resources, tangible or virtual, that
accrue to a corporate player through the players
social relationships, facilitating the attainment
of goals. (Leenders Gabbay 19993)
3
CSC through SANs
A firms ties to organizations in a strategic
alliance network increases its probability of
accessing and using the valuable CSC resources
held by the firms partners, including their

• Financial resources, credit extensions
• Knowledge, information, technologies/patents
• Marketing expertise, country/culture penetration
• Orgl statuses, corporate/brand reputations
• Trustworthiness and low risk (moral hazards)

Organizations aware of such CSC advantages may
act strategically in pursuing new alliances,
partnering with firms that maximize its CSC
portfolio. At the field-net level, an evolving
strategic alliance network comprises a collective
CSC structure which simultaneously facilitates
and constrains the opportunities for its member
firms.
4
Strategic Alliance Networks
Corporate social capital relations span multiple
levels of analysis from individuals, to
workteams, to firms, and organizational field
network (Kenis Knoke 2002). At the IOR level,
repeated alliances generate a strategic alliance
network form of CSC.
Strategic alliance at least two partner firms
that (1) remain legally independent (2) share
benefits, managerial control over performance of
assigned tasks (3) make contributions in
strategic areas, e.g., technology or products
(Yoshino Rangan 1995).
Strategic alliance network The set of
organizations connected through their overlapping
partnerships in different strategic alliances
(Knoke 2001128 Todeva Knoke 2002). Firms are
closely tied to one another through many direct
alliances or many indirect ties through third
firms (i.e., partners-of-partners).
5
Global Information Sector
Basic CSC concepts could help to explain the
evolution of the strategic alliance network in
the Global Information Sector (GIS). This sector
increased collaborative agreements exponentially
1989-2000, creating a complex web of overlapping
partnerships.

• Five NAICS info subsectors (publishing motion
  pictures sound recording broadcasting
  telecomms info services data processing) plus
  computer, telecomm, semiconductor manufacturing
  industries
• 145 multinational corporations 66 USA, 16
  Europe, 15 Asia
• Alliance venture announcements in general
  business news media from 1989 to 2000
• Total of 3,569 alliances involving two or more
  GIS organizations (some alliances include noncore
  partners)

6
Research Hypotheses
Three types of Hs about network evolution
involve changes in global structure, partner
choice, and organizational performances.
H1 Network Structural Change The GIS SAN
evolved from a fragmented small world of
specialized cliques toward preferential
attachments to key producers, and then to
structurally cohesive connectivity.
H2a Transitivity Firms are more likely to form
new alliances with other organizations that
result in transitivity. H2b Balance Firms with
a specific number of partners are more likely to
form new alliances with other orgs having an
identical or very similar N of partners. H2c
Indirect Relations Firms are more likely to form
new alliances with other organizations to which
they are linked by numerous indirect
connections. H2d Similarity / Interdependence
Firms are more likely to form new alliances with
other organizations that having similar /
complementary attributes.
7
Rising Alliance Rates
8
Diverse Purposes
9
Closeness Centrality
CENTRALITY ORGS INVOLVED WITH MANY
PARTNERS DEGREE Number of ties directly
connecting focal org to other orgs (in- or
out-degrees) CLOSENESS Inverse of sum of
distances to other orgs (geodesics shortest
paths) NETWORK CENTRALIZATION Extent to which
one actor has high centrality and others low
CLOSENESS 1991 ATT 1995 IBM Sun Intel 2000
Microsoft IBM Sun HP
10
Betweenness Centrality
CENTRALITY ORGS INVOLVED WITH MANY
PARTNERS BETWEENNESS Number of times an org
occurs on a geodesic between other pairs of
orgs NETWORK CENTRALIZATION Extent to which one
actor has high centrality and others low
BETWEENESS 1991 ATT Time Warner 1995 ATT
Intel IBM 2000 Microsoft IBM
11
MAPPING The GIS CORE
Hierarchical cluster multidimensional scaling
analyses to identify positions and spatial
proximities among 30 most-active GIS firms (1991,
1995, 2000). Similarity N of partnerships
per dyad.
12
(No Transcript)
13
SAMSUNG
14
(No Transcript)
15
(No Transcript)
16
Evolution Analysis
The macro-evolution of GIS alliance network,
under dynamic constraints of network properties,
assumes methodological individualism
(actor-oriented model) SIENA (Simulation
Investigation for Empirical Network Analysis
Snijders 2005) models the changing network
connections as outcomes of orgl decisions to add
or drop ties, assuming that orgs seek to maximize
various objective function elements (e.g.,
preferences for increased network transitivity,
reciprocity, balance, alliances with popular and
active partners, etc.)
SIENA estimates effects using two or more
observed matrices of dichotomous ties. It
applies the method of moments, implemented as a
continuous-time Markov chain Monte Carlo
simulation (MCMC) i.e., actors know networks
current structure, but not its earlier states.
17
GIS Core Firm Alliances
SANs among 26 GIS firms 1998-99-00 (binarized at
2 per year). Here is the 2000 matrix, density
0.618
18
(No Transcript)
19
Evolution of the GIS Core
SIENA analysis of strategic alliances
(dichotomized at 2 per year) among the 26
most-active GIS firms for 1998-1999-2000.
Results consistent with all H2s except
transitivity hypothesis.
OBJECTIVE FUNCTION Parameter Stnd error t-ratio
Rate (1998-1999) 11.82 2.67 4.43
Rate (1999-2000) 8.41 1.95 4.31
Density (degree) 0.71 0.12 5.92
Transitivity 0.01 0.07 0.14
Balance 1.34 0.31 4.32
Indirect Relations 0.69 0.19 3.63
Geographic Similarity 0.47 0.13 3.62
Industry Similarity 0.21 0.19 1.11
p lt .05 p lt .01 p lt .001
20
Issues in SAN Evolution
? What substantive interpretations can we make
about the SIENA parameters? How robust for the
larger GIS network and longer evolutionary span?
? Which, if any, tie-formation processes in
interorganizational relations are functionally
equivalent to interpersonal choices? ? Do
balance and transitivity have the same meanings
in organizational partnering and social
psychological affiliation? ? Are different
theoretical concepts, principles, and
propositions necessary to explain
interorganizational network dynamics? If so,
what are they?
21
Further Steps
GIS orgs built up extensive corporate social
capital by rapidly expanding the worldwide
strategic alliance network. Structural cohesion
seems increasing important for collective actions
and individual firm outcomes. By expanding the
GIS dataset to cover 1986-2005, I hope to track
transformations in structures and processes from
the Sectors origins to well beyond the bursting
of the Dot.com Bubble in Spring 2000. Using
data on firm profits, growth, patent innovations,
I will test the third set of hypothesis about
organizational performance Are structurally
equivalent or socially cohesive clusters of
collaborating organizations better able to use
the structural advantages of jointly occupied
network positions to access valuable information,
obtain scarce resources, and improve their
members performances?
By helping to provide policymakers with a deeper
understanding of the types of alliance networks
that affect firm innovations, subsequently
modified legislative, regulatory, and trade
association policies might be crafted to foster
the development of interorganizational
connections with optimal structural
characteristics.
22
References
Burt, Ronald S. 1992. Structural Holes The
Social Structure of Competition. Cambridge, MA
Harvard University Press. Coleman, James S.
1990. Social Capital. Pp. 300-321 in
Foundations of Social Theory. Cambridge, MA
Harvard University Press. Kenis, Patrick and
David Knoke. 2002. How Organizational Field
Networks Shape Interorganizational Tie-Formation
Rates. Academy of Management Review
27275-293. Knoke, David. 2001. Changing
Organizations Business Networks in the New
Political Economy. Boulder, CO
Westview. Leenders, Roger Th. A. J. and Shaul M.
Gabbay (eds.). 1999. Corporate Social Capital and
Liability. Boston Kluwer Academic
Publishers. Lin, Nan. 2001. Social Capital A
Theory of Social Structure and Action. New York
Cambridge University Press. Snijders, Tom A.B.
2005. Models for Longitudinal Network Data. Pp.
215-247 in Models and Methods in Social Network
Analysis, edited by Peter J. Carrington, John
Scott and Stanley Wasserman. New York Cambridge
University Press. Todeva, Emanuela and David
Knoke. 2002. Strategische Allianzen und
Sozialkapital von Unternehmen. (Strategic
Alliances and Corporate Social Capital) Kölner
Zeitschrift für Sociologie und Sozialpsychologie.
Sonderheft 42345-380. Yoshino, Michael Y. and
U. Srinivasa Rangan. 1995. Strategic Alliances
An Entrepreneurial Approach to Globalization.
Cambridge, MA Harvard University Press.